Cnv.clustered_barcodes

missionbio.mosaic.cnv.Cnv.clustered_barcodes

Cnv.clustered_barcodes(orderby: Optional[Union[str, ndarray, DataFrame]] = None, splitby: Union[str, ndarray, DataFrame] = 'label', override: bool = False, subcluster: bool = True, return_index: bool = False, **kwargs: Any) → ndarray

Hierarchical clustering of barcodes.

If orderby is not provided, the ordered barcodes are returned according to splitby. If orderby is provided, then further hierarchical clustering is performed based on the values in it for each unqiue splitby value.

Parameters:

orderbystr / np.ndarray

An attribute with at least one dimension equal to the number of barcodes. Uses get_attribute() to retrieve the values constrained by row.

splitbystr / np.ndarray

Whether to order the barcodes based on the given labels or not. Only applicable when labels orderby is None. Uses get_attribute() to retrieve the values constrained by row. The shape must be equal to (#cells)

overridebool

Continue clustering even if the number of id is greater than 1,000.

subclusterbool

Whether to hierarchically cluster the barcodes within each unique splitby value cluster.

return_indexbool

Whether to return the indices of the ordered barcodes or the barcodes themselves.

kwargsAny

The keyword arguments to be passed to the scipy.cluster.hierarchy.linkage() function. See hierarchical_cluster() for the default values.

Returns:

numpy.ndarray

Ordered barcodes

Raises:

ValueError

When the splitby is not one-dimensional

Exception

When the number of id in the assay is greater than 1,000. This can take a long time to process, hence denied.

Warning

If the number of id is greater than 1,000 and override is passed.

---

< Class Cnv